MODELS SCALE AND THE ENIGMA OF THE STEWARD As knowledge has advanced in the world has gradually become clear that the everyday objects or physical systems are in turn composed of smaller parts that are the ; affect their behavior. To not have been considered minor parties that make up matter had been left out of any model of the world a considerable amount of information that has led to a loss of predictive ability.
If set
last elements to those elements that contain all the information of a physical system, then for a physical system with information provided last elements must exist. In the physical world there are some basic quantities such as speed, electric charge, mass, etc. The information that is to label all parts under one system is greater than the result of joint label formed by the parties, so the last element must be the result of fewer than one system.
UNCERTAINTY uncertainty is defined here as the degree of deviation that has a certain calculation or prediction of a model with respect to the real situation they try to emulate. For example, a broker dedicated to technical analysis of charts of the bag has a high uncertainty because their forecasts are barely better than chance.
therefore to establish the uncertainty of a model could be based in the following scheme:
- Calculate the results x ^ m which gives the model for all situations i in which prentede is applied. Access
- a record of all results x ^ r for all situations i and compared with those obtained with the model. Calculate
- half or a statistical distribution of the different measures.
can propose a mathematical definition of uncertainty such as the following:
strict causality strict causality is understood when the elements and operations from which defines the system under study to simulate with the uncertainty that emulate the physical system. For example if the universe is defined as classical mechanics, the universe would have no uncertainty and serious classical mechanics strict causality.
A strict causal model should incorporate in their model the last elements of the physical world, otherwise there would be information that would not have been contemplated. and can affect the evolution of the system. If this information loss does not affect the evolution of the physical system, it has no physical meaning and can be asserted in fact they have been taken into account the last elements to build the model. According to current science, the last element would subatomic particles. PARTIAL CAUSATION
partial causality is understood when the elements and operations from which defines the system under study to simulate with greater uncertainty than the physical system itself emulate. Within this group there are models ranging from a low uncertainty or high uncertainty, models that provide knowledge and models that contribute less or nothing at all.
SCALE OF RANDOMNESS
Suppose all experiments that can see a model and actual yield values \u200b\u200bXi, for any model whose uncertainty is greater than 0 and Xi seeks to calculate the exact figures are significant in that calculation with equal certainty to establish those numbers at random, so there is a random zone around Xi. Put simply, there are decimals in the calculation made by a model that are disposable because not manage more than chance. That is, in these cases there is a scale in which the model provides no information, this is the range of validity of the model. This scale will be greater or less depending on the model uncertainty.
If Xi are known and all results are superimposed on the same line at a point X, the chance is one area comprising half of the calculated values \u200b\u200bcloser to Xi, given that the border drawn through these points divided by 50% the results of different experiments.
TECHNICAL LIMITATIONS
At a time given the state of the art there is limited capacity for obtaining data from a system that seeks to emulate and a limited ability to evaluate these data computation according to a model.
MODELS FOR SCALES
For example, in the scale of the last items given as prior art that emulates the model can not do it on a random high number of elements, there will come a time that is needed for a new model which used some simplifications based on new definitions and standards to emulate a larger system, eliminating information that is entered in the assessment. The removal of this information should bring an increase in model uncertainty necessarily on the knowledge of the state last elements of physics and predictably knowledge about the state of the new elements conjugate.
According to the figure, although the theory being strict certainty to the practicability is limited to a sample small. At which a theory B which includes some simplifications and approximations is practicable for a larger number of elements, either because keeping the amount of information has decreased particle density and / or because the state of the art allows better access the information contained in the new conjugated elements. This theory B a C etc can happen.
MORE CHANGE SCALE MODEL Suppose that information on the working model has the form of a vector (a, b, c. ..... n) for each element considered. Since the number of vectors would be limited and ordering these vectors would have a finite dimensional matrix. On this matrix the model performs a transformation into a resulting array. For example:
To create a larger-scale model using the previous vectors and are linearly combined (or not) for new vectors representing elements of the new model compounds. Also combining the columns that represent the new variables of this new model of larger scale. For example, the temperature variable is a combination of the columns for mass and velocity, which makes sense in a model that has also made a combination of a large number of rows to result in much less, making the total size of the matrix and the information it contains is smaller.
The result of successive iterations is that the number of rows in the matrix tends to decrease as the generation of new columns tends to increase.
THERE FOR ALL YOUR MODEL MATRIX SCIENTIFIC may seem that at a certain time of the change of models to a larger scale need to leave the combinations of columns to find new variables could be flexible enough such
" species, mountain, pulley ", however currently the voice recognition software and faces do well in their role and are based on mathematically combine and process basic physical information (color, position, pressure).
is a sufficient argument follows: since any physical system capable of doing science is ultimately reducible to its last element array, if A is the family of all sets of vectors that the system capable of understanding science as a scientific concept then there an equation whose result is 0 when evaluated solely on any of these sets of vectors that the physical system capable of doing science as a concept means. What's more, for each concept whether scientific or otherwise there is a mathematical combination that wins his column in the matrix of variables.
PILGRIM CLASSIFICATION BY SCALE MODELS IN THE NEWS Under the provision that a larger-scale theory is adjacent to another of the parties immediately to get simpler than its elements are conjugated, missing information in the process. Can venture a classification to vote soon on some of the branches of knowledge about nature.
